Pathology Visual Question Answering on PathMMU (test-all)
77.1Overall ScoreGemini-3-Pro-Preview
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Gemini-3-Pro-Preview2026.05 | 77.1 | 79 | 74.5 | 77.5 | 85.4 | 69.2 | |
| PathoSageHost Model=GPT-5.42026.05 | 76.8 | 77.1 | 76.5 | 79.2 | 82.6 | 71 | |
| Gemini-3-Flash-Preview2026.05 | 76.5 | 78.8 | 74.5 | 77.2 | 83.8 | 66.1 | |
| CPath-Omni2026.05 | 72.2 | 69.9 | 71.8 | 70.6 | 70.6 | 79 | |
| GPT-5.42026.05 | 70.8 | 74.2 | 71.2 | 75.4 | 75.1 | 55.5 | |
| Patho-AgenticRAG2026.05 | 70.5 | 71 | 72.9 | 73.8 | 78.8 | 55.2 | |
| GPT-5.4-mini2026.05 | 66.5 | 69.9 | 68.9 | 71.6 | 68.5 | 50.4 | |
| Patho-R12026.05 | 66.5 | 69.2 | 67.9 | 70.9 | 78.5 | 45 | |
| Qwen3-VL-32B-Instruct2026.05 | 61.9 | 64.2 | 65.5 | 65.4 | 66.8 | 49.4 | |
| PathGen-LLaVA2026.05 | 58.4 | 60.1 | 58.8 | 60.7 | 64.9 | 48.9 | |
| HuatuoGPT-7B2026.05 | 56.4 | 61.7 | 58.6 | 57.4 | 62 | 38.4 | |
| Qwen3-VL-8B-Instruct2026.05 | 54.8 | 59.4 | 58.5 | 55.6 | 55.6 | 42.6 | |
| InternVL3-8B2026.05 | 52.9 | 55.9 | 57 | 52.9 | 56.2 | 41 | |
| InternVL2.5-8B2026.05 | 48.6 | 53 | 54.4 | 50.6 | 50.8 | 31.1 | |
| Quilt-LLaVA2026.05 | 41.5 | 42.6 | 46.6 | 45.3 | 42.7 | 29.2 | |
| LLaVA-1.5-13B2026.05 | 37.6 | 41 | 40.4 | 39.4 | 44.3 | 23.5 | |
| LLaVA-Onevision-7B2026.05 | 34.4 | 36.4 | 38.4 | 38.3 | 38.4 | 20.4 | |
| InstructBLIP-FLAN-T5-XXL2026.05 | 33.9 | 37.2 | 34.3 | 36 | 39.3 | 22.7 | |
| LLaVA-Med2026.05 | 26.2 | 27.7 | 27.3 | 27.2 | 30.7 | 20.3 |